A user-friendly ML platform that lets non-experts upload datasets, receive model suggestions, schedule training, and deploy models for real-time predictions
0
stars
0
forks
source link
ML & Deployment- Package training job into runnable Docker #29
Users need to be able to schedule model training easily through out app. In order to utilize the developed python scripts remotely, we need to run them through a server so that an end user can access them anywhere no matter the level of compute power is on their system.
Persona:
Denzel does not have access to a GPU accelerated system locally, and would like to be able to run his models in a faster timeframe no matter where he is. In order to do this, we will need to host the app and software on a remote server.
Feature:
Provide access to model script endpoints and data.
Business Value:
Users will be able to run training jobs with simple button clicks due to dockerized scripts.
Tasks
Current implementation plan is to develop a docker script that can run the python interpreter with associated required libraries.
Acceptance Criteria
Dockerfile script should be runnable via command line commands and should output the results from running a model type function on the ML API.
Acceptance Tests
Compile and run docker file via command line with specific flags to utilize functions in ML API class.
User Story:
Users need to be able to schedule model training easily through out app. In order to utilize the developed python scripts remotely, we need to run them through a server so that an end user can access them anywhere no matter the level of compute power is on their system.
Persona:
Denzel does not have access to a GPU accelerated system locally, and would like to be able to run his models in a faster timeframe no matter where he is. In order to do this, we will need to host the app and software on a remote server.
Feature:
Provide access to model script endpoints and data.
Business Value:
Users will be able to run training jobs with simple button clicks due to dockerized scripts.
Tasks
Current implementation plan is to develop a docker script that can run the python interpreter with associated required libraries.
Acceptance Criteria
Dockerfile script should be runnable via command line commands and should output the results from running a model type function on the ML API.
Acceptance Tests
Compile and run docker file via command line with specific flags to utilize functions in ML API class.